This process mining project discovers the process model of an event log using the process mining algorithms Alpha Miner, Heuristics Miner and Inductive Miner.
After discovering the process models, it performs evaluation, finding it's fitness, precision, generalization and simplicity values.
Lastly it performs conformance checking using the Replay Fitness method. Conformance checking compares the discovered model with the actual event log to identify possible deviations and potential bottlenecks.
The event log contains activities of a smart home.
This project was made during my Intelligent Systems course in University.
- The event log file needs to be in the same directory as the python script file.
- Execute
py <filename>.py.
Alpha Miner Process model
Heuristics Miner Process model
Inductive Miner Process model
| fitness | precision | generaliztion | simplicity | |
|---|---|---|---|---|
| Alpha miner unfiltered log | 0.38 | 0.02 | 0.89 | 1.0 |
| Alpha miner filtered log | 0.66 | 0.02 | 0.89 | 1.0 |
| Heuristic miner unfiltered log | 0.95 | 0.31 | 0.69 | 0.51 |
| Heuristic miner filtered log | 0.94 | 0.31 | 0.71 | 0.52 |
| Inductive miner unfiltered log | 0.98 | 0.15 | 0.87 | 0.64 |
| Inductive miner filtered log | 0.98 | 0.15 | 0.88 | 0.64 |